Voicebox — локальная студия синтеза и клонирования голоса

★ 47.5K

Voicebox is an open-source voice studio that runs locally: voice cloning, text-to-speech and dictation in one application. Reach for it when you want to narrate a video, a course or a podcast in your own voice without sitting down at a microphone, voice material in several languages, or simply dictate text instead of typing. It works in both directions — text to speech and speech to text. Seven synthesis engines are wired in — Qwen3-TTS, Qwen CustomVoice, LuxTTS, two Chatterbox variants, HumeAI TADA and Kokoro — covering 23 languages with over fifty ready voices, and cloning works from a few seconds of audio. You can switch engines in the interface and compare how each sounds on your own text. Everything is computed on your machine and no recordings are sent anywhere.